Synthesis of 4-haloserotonin derivatives and synthesis of the toad alkaloid dehydrobufotenine

Article abstract of DOI:10.1016/j.tet.2010.04.081

4-Chloro-, 4-bromo- and 4-iodoserotonin derivatives were synthesized from 4-halo-5-oxyindoles, themselves derived from coumarins. Synthesis of the 4-iodoserotonins involved conjugate addition of an allyl group to a 5-iodocoumarin, followed by conversion of the allyl group into an aminoethyl unit. One of the iodoserotonin derivatives was converted into the toad alkaloid dehydrobufotenine, which was isolated as its hydroiodide.

Full text of DOI:10.1016/j.tet.2010.04.081

Tetrahedron 66 (2010) 4452e4461
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Synthesis of 4-haloserotonin derivatives and synthesis of the toad alkaloid
dehydrobufotenine
*
Elia J.L. Stoffman, Derrick L.J. Clive
Chemistry Department, University of Alberta, Edmonton, Alberta, Canada T6G 2G2
a r t i c l e i n f o
a b s t r a c t
Article history:
4-Chloro-, 4-bromo- and 4-iodoserotonin derivatives were synthesized from 4-halo-5-oxyindoles,
themselves derived from coumarins. Synthesis of the 4-iodoserotonins involved conjugate addition of an
allyl group to a 5-iodocoumarin, followed by conversion of the allyl group into an aminoethyl unit. One of
the iodoserotonin derivatives was converted into the toad alkaloid dehydrobufotenine, which was iso-
lated as its hydroiodide.
Received 20 March 2010
Received in revised form 19 April 2010
Accepted 19 April 2010
Available online 24 April 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
4-Haloserotonins
Dehydrobufotenine
Coumarins
Haloindoles
1. Introduction
X
X
R
conjugate
addition
MsO
HO
CO2Et
O
MsO
5-Hydroxytryptamine (serotonin) and some of its derivatives
have a wide range of important biological properties.1,2 A poten-
tially useful synthetic route to analogues of 5-hydroxytryptamine
would involve palladium-mediated coupling reactions of various
halo tryptamine derivatives.3 Among these derivatives, one sub-
groupdthose with a halogen at C(4) of the indole nucleusdis not
easily accessed by current methodology.4 Such compounds may be
useful in routes to complex indoles4c and could also be medically
important because some 4-haloserotonins are reported in patent
literature4b that deals with the development of drugs to treat mi-
graine and portal hypertension.
decarboxylation
O
O
1.2
O
1.1 X = Cl, Br, I
NH3, THF; O-protection
Pb(OAc)4, t-BuOH
X
R
X
R
O-deprotection
MsO
NHBoc
oxidation
N-deprotection
N
H
OPg
isomerization
1.4
1.3 Pg = protecting
group
Scheme 1. General route for conversion of 5-halocoumarins into indoles.
2. Results and discussion
ring to a quinone. The N-protecting group is then removed and
cyclization occurs to form a quinone imine, which isomerizes to the
indole, either spontaneously or on treatment with a catalytic
amount of RheAl2O35 or RheC. The nitrogen protecting group (Boc)
can also be removed prior to oxidizing the benzene ring.
The two simplest routes to the serotonins we have made are
those leading tothe chloro compound 2.8 (Scheme 2) and the bromo
analogue 3.8 (Scheme 3). The chloro series begins with silylation of
indole 2.1, which was made by the method we have developed,5
followed by VilsmeiereHaack formylation, to produce aldehyde
2.3 (2.1/2.2/2.3). The nitrogen was then protected by sulfonyla-
tion, and base catalyzed condensation with MeNO2 gave the
unsaturated nitro compound 2.5. At this stage the pendant double
bond was saturated by treatment with LiBH4 (76%)7 and the nitro
We have recently developed a method for making 5-hydroxy-4-
haloindoles5 from readily accessible coumarins, and we now
describe the application of this method to the preparation of
a number of 4-haloserotonin derivatives and to the synthesis of the
toad alkaloid6 dehydrobufotenine.
The general principle of our approach5 to 5-hydroxy-4-halo-
indoles (Scheme 1) involves ring opening of dihydrocoumarins
with ammonia, followed by Hofmann rearrangement of the
resulting amide to an N-Boc amine, and oxidation of the aromatic
* Corresponding author. Tel.: þ1 780 492 3251; fax: þ1 780 492 8231; e-mail
address: derrick.clive@ualberta.ca (D.L.J. Clive).
0040-4020/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.04.081

E.J.L. Stoffman, D.L.J. Clive / Tetrahedron 66 (2010) 4452e4461
4453
group was reduced with ZneAcOH. Finally, N-acetylation yielded
the chloroserotonin derivative 2.8. The corresponding bromoser-
otonin derivative 3.8 was made in exactly the same way (Scheme 3).
In both the above routes the zinc reduction (i.e., 2.6/2.7 and
3.6/3.7) was completely selective for the aliphatic nitro group
over the aromatic halogen; however, this method cannot be applied
to the iodo series because the iodine is lost during the Zn reduction
step. Consequently, the route described below (Scheme 4) was
developed in order to gain access to 4-iodoserotonins, which we
expected to be more useful for subsequent transition metal-me-
diated coupling reactions.8
t-BuMe2SiCl
Cl
Cl
ImH, CH2Cl2
76%
HO
Si*O
Si*O
Si*O
t-BuMe2SiO
N
H
N
H
ROH, Ti(OPr-i)4
2.2
2.1
PhMe, reflux;
4.2a (88%)
4.2b (-%)
I
I
POCl3, DMF
Cl
MsO
CO2Et
O
MsO
CO2R
O
Cl
O
CHO
CHO
O
TsCl, Et3N, DMF
Si*O
4.1
4.2a R = CH2CH2SiMe3,
63% from 2.2
4.2b R = allyl
allylmagnesium
bromide, CuI, LiCl
Me3SiCl, THF, -40 °C
4.3a 99%
4.3b 92% from
4.1
N
N
H
SO2Tol
2.4
2.3
NH4OAc, MeNO2
Si* = SiMe2Bu-t
Bu4NF, AcOH
reflux
I
I
DMF, 100 °C
96%
NO2
NO2
RO
MsO
CO2R
Cl
Cl
LiBH4, THF, 0 °C
Si*O
76% from 2.4
O
O
O
O
MsCl
Et3N
4.4a R = H,
4.4b R = Ms
4.3a
N
N
4.3b
CH2Cl2
72%
AcOH, Pd(PPh3)4,
Ph3P, CH2Cl2,78%
SO2Tol
SO2Tol
2.6
2.5
NH3, THF
91% from
4.4b
Zn, AcOH
dioxane, rt
allyl bromide
K2CO3
NHAc
NH2
I
I
Cl
Cl
2-butanone
Ac2O, pyr, THF
MsO
CONH2
MsO
CONH2
Si*O
60 °C, 95%
35% from 2.6
N
N
OH
OAllyl
4.5
4.6
SO2Tol
SO2Tol
2.8
2.7
Pb(OAc)4 80 °C, 91%
Scheme 2. Synthesis of a chloroserotonin derivative.
t-BuOH
Triton B, aq
dioxane
45 °C
I
I
HO
NHBoc
NH3Cl
MsO
NHBoc
OAllyl
t-BuMe2SiCl
92%
Br
Br
ImH, CH2Cl2
HO
Si*O
Si*O
Si*O
t-BuMe2SiO
80%
OAllyl
4.7
4.8
HCl
N
N
H
PhI(OCOCF3)2
CF3CO2H
aq MeCN
0 °C
H
EtOAc
99%
I
3.1
3.2
I
POCl3, DMF
(inverse
O
HO
addition)
Br
Br
CHO
CHO
N
OAllyl
Si*O
TsCl, Et3N, DMF
4.9
4.10
I
Rh/C, PhH
reflux
RO
N
N
H
SO2Tol
3.4
3.3
N
H
NH4OAc, MeNO2
Si* = SiMe2Bu-t
reflux, 53% from 3.2
t-BuMe2SiCl, ImH
4.11 R = H
4.12 R = SiMe2Bu-t
NO2
NO2
CH2Cl2, 67% from 4.9
Br
Br
LiBH4, THF
Scheme 4. Synthesis of 3-allyl-5-hydroxy-4-iodoindole.
Si*O
0 °C, 98%
N
N
The method, which we used to prepare the starting materials
2.15 and 3.15 for the present work can be easily modified5 so that it
affords indoles carrying a substituent at C(3), and we now took
advantage of this useful feature to make 4-iodoserotonin de-
rivatives. We chose an allyl group as the C(3) substituent so that
oxidative cleavage of the double bond would give an aldehyde that
could undergo reductive amination, using conditions that are
compatible with the presence of an aromatic iodide.
SO2Tol
SO2Tol
3.5
3.6
Zn, AcOH
dioxane, rt
NHAc
NH2
Br
Br
Ac2O, pyr, CH2Cl2
Si*O
93% from 3.6
N
N
Coumarin 4.15 was subjected to ester exchange with 2-(trime-
thylsilyl)ethanol and also with allyl alcohol (Scheme 4). This ex-
change process was necessary because use of the ethyl ester did not
SO2Tol
SO2Tol
3.8
3.7
Scheme 3. Synthesis of a bromoserotonin derivative.

4454
E.J.L. Stoffman, D.L.J. Clive / Tetrahedron 66 (2010) 4452e4461
allow reproducible or efficient hydrolysis and decarboxylation at
a later stage (see 4.3a,b/4.4a,b). Conjugate addition to the new
esters, using allylmagnesium bromide and CuI in the presence of
LiCl and Me3SiCl,9 afforded the dihydrocoumarins 4.3a (99%) and
4.3b (92% from 4.1), respectively. The next step was a de-
carboxylation for which the method used depended on the nature
of the ester. In the case of 4.3a, Bu4NFeAcOH in hot DMF (100 ꢀC)
was satisfactory, but the methanesulfonyl group was lost at the
same time and had to be replaced (4.3a/4.4a/4.4b). The high
cost of 2-(trimethylsilyl)ethanol and the need to replace the
methanesulfonyl group (subsequent reaction with NH3 did not
work with a free hydroxyl) made us examine the allyl ester series.
Decarboxylation of allyl ester 4.3b was first effected with Pd(PPh3)4
in the presence of an excess of dimedone (4.3b/4.4b, 61%), but we
later found that AcOH and catalytic amounts of Ph3P and Pd(Ph3P)4
in refluxing CH2Cl2 was much superior, since with this reagent
system10 all byproducts are volatile and the yield was higher (78%).
The mesylate 4.4b was then subjected to the following sequence:
lactone opening with NH3, O-allylation and Hofmann rearrange-
ment (4.4b/4.5/4.6/4.7). At this point, Triton B and HCl were
used to remove the methanesulfonyl group and the nitrogen pro-
tecting group, respectively (4.7/4.8/4.9). Oxidation [PhI
(OCOCF3)2] and isomerization of the resulting quinone imine (4.10),
using a catalytic amount of 5% RheAl2O3, then gave the 3-ally-
lindole 4.11. The oxidation step required some optimization to avoid
formation of a purple byproduct. Use of PhI(OCOCF3)2 in the pres-
ence of CF3CO2H11 was superior to PhI(OAc)2 alone11 and, to sup-
press formation of the purple byproduct, the quinone imine 4.10
and the final indole (4.11) were protected from light, and a trace of
2,6-di-tert-butyl-4-methylphenol (BHT) was added to their solu-
tions. Evaporation of solutions of these compounds to dryness was
only done immediately before changing the solvent and further
processing (see Experimental section). We found it best to prepare
4.10 and 4.11 the same day as they were to be used. O-Silylation of
4.11 (4.11/4.12) made the compound much more stable.
In the present case it was treated with BnNH2 and NaBH3CN14 to
produce the iodoserotonin derivative 5.4.
As a simple example to demonstrate the use of the iodoserotonin
5.4 in synthesis, we converted the compound into the hydroiodide
of the toad alkaloid dehydrobufotenine (6.4).6,15 This was done
(Scheme 6) by palladium-mediated intramolecular amination
(5.4/6.1), along the lines of a reported15b,16 method. The use of t-
BuONa16 rather than K2CO315b was essential for our particular sub-
strate, in contrast to a related15b cyclization where K2CO3 was the
required base. Removal of the sulfonyl group with Na(Hg) in dry
17
MeOH buffered with Na2HPO4 (6.1/6.2), followed by hydro-
genolysis of the benzyl group (6.2/6.3) and dimethylation with
MeI, gave dehydrobufotenine hydroiodide (6.4) directly, since the
silicon protecting group was conveniently lost during the methyl-
ation. In preliminary experiments, we had removed the silicon
protecting group from 6.2, with the intention of then removing the
benzyl group; however, the phenol resulting from desilylation of 6.2
was unstable and decomposed rapidlyda property that dictated the
order of the last few steps that led to the alkaloid (Scheme 6).
NHBn
Bn
Pd(PPh3)4, t-BuONa
I
N
Et3N, PhMe
Si*O
Si*O
200 °C, 18 h, 43%
N
N
SO2Ph
SO2Ph
6.1
5.4
Si* = SiMe2Bu-t
Na(Hg), Na2HPO4
MeOH, 98%
H
Bn
N
N
Pd/C, H2
EtOH, ca 77%
Si*O
Si*O
N
N
H
H
6.3
6.2
I-
+
MeI
N
The allylindole 4.12 was then converted (Scheme 5) into the
iodoserotonin derivative 5.4. To this end, the nitrogen of 4.12 was
protected by sulfonylation12 (4.12/5.1). Cleavage of the pendant
double bond12 by dihydroxylation (OsO4) and treatment with NaIO4
on silica gel13 gave aldehyde 5.3, which we expect will be a versatile
intermediate for the preparation of serotonin derivatives since it
should undergo reductive amination with a wide range of amines.
Na2CO3
69%
HO
N
H
6.4
Scheme 6. Synthesis of dehydro bufotenine.
3. Conclusion
PhSO2Cl, KOH
Bu4NHSO4
PhH
I
I
The method reported here for the synthesis of 4-haloserotonins,
especially the iodo compound, should be useful for a variety of
coupling reactions that would give access to a large range of sero-
tonin derivatives. The ability to incorporate an allyl group at C(3) of
the indole nucleus gives easy access to iodoserotonin derivatives
because it allows introduction of an amino group under reductive
conditions that are fully compatible with the presence of the iodine
atom; this reductive amination should be general for numerous
primary and secondary amines.
Si*O
Si*O
Si*O
Si*O
N
H
N
SO2Ph
5.1
4.12 Si* = SiMe2Bu-t
OsO4, NMO
THF, H2O
63% from 4.12
OH
I
CHO
I
NaIO4-silica
CH2Cl2
OH
Si*O
N
N
4. Experimental section
4.1. General
SO2Ph
SO2Ph
5.3
5.2
BnNH2, AcOH, NaCNBH3
MeOH, THF, 72% from 5.2
NHBn
Column sizes are quoted in the form diameterꢁlength. Evapo-
rations were done under water-pump vacuum using a rotary
evaporator and the residue was then kept under oil-pump vacuum.
Ar and N2 were purified by passage through a column (3.5ꢁ42 cm)
of BASF R-311 catalyst and then through a similar column of
Drierite. Glassware was dried in an oven (120 ꢀC) for at least 3 h
before use and either cooled in a desiccator over Drierite, or
I
N
SO2Ph
5.4
Scheme 5. Synthesis of a 4-iodoserotonin derivative.

E.J.L. Stoffman, D.L.J. Clive / Tetrahedron 66 (2010) 4452e4461
4455
assembled quickly, sealed with rubber septa and allowed to cool
under a slight static pressure of Ar or N. Reaction mixtures were
stirred by Teflon-coated magnetic stirring bars. Solvents for chro-
matography or extractions were distilled before use. Melting points
were determined on a Kofler block melting point apparatus. Com-
mercial thin layer chromatography (TLC) plates (silica gel, Merck
60F-254) were used. Compounds were detected by examination
under UV light or by dipping the plate into a solution of phos-
phomolybdic acid (Ref. 18 in Supplementary data), followed by
charring with a heat gun. Silica gel for flash chromatography was
Merck type 60 (230e400 mesh). Dry solvents were prepared under
an inert atmosphere and transferred by oven-dried syringes. Dry
THF was distilled from Na and benzophenone ketyl. Dry PhH and
PhMe were distilled from Na. Dry Et3N, CH2Cl2, MeOH, pyridine and
DMF were distilled from CaH2, the last two solvents being distilled
under water-pump vacuum. The symbols s, d, t and q used for 13C
NMR spectra (ATP) indicate 0, 1, 2 or 3 attached hydrogens. 1H NMR
spectra were recorded with Varian INOVA-300 (at 300 MHz), Var-
ian INOVA-400 (at 400 MHz) and Varian INOVA-500 (at 500 MHz)
spectrometers in the specified deuterated solvent. 13C spectra were
recorded with Varian INOVA-400 (at 100 MHz) and Varian INOVA-
500 (at 125 MHz). Mass spectra were recorded with Agilent Tech-
nologies 6220 Accurate-Mass TOF LC/MS, Perseptive Biosystems
Mariner Biospectrometry Workstation, Kratos MS50 or Micromass
ZabSpec Hybrid Sector-TOF mass spectrometers.
3.3 had: FTIR (film cast) 3108, 2928, 2894, 2856, 1626, 1570, 1504,
1463, 1429, 1386, 1334, 1268, 1252, 1232, 1080, 983, 920, 838, 799,
781, 667 cmꢂ1 1H NMR (500 MHz, CDCl3) (one H not visible in
;
spectrum)
d
0.27 (s, 6H), 1.08 (s, 9H), 6.70 (d, J¼8.6 Hz, 1H), 7.30 (d,
J¼8.6 Hz, 1H), 8.05 (d, J¼2.9 Hz, 1H), 9.08 (br s, 1H); 13C NMR
(125 MHz, CDCl3)
d 0.0 (q), 22.5 (s), 29.9 (q), 109.8 (s), 115.7 (d),
120.9 (d), 124.1 (s), 130.8 (s), 135.3 (d), 136.4 (s), 153.0 (s), 191.1 (d);
exact mass (electrospray) m/z calcd for C15H2079BrNNaO2Si (MþNa)
376.0339, found 376.0337.
4.4. 4-Bromo-5-[(tert-butyldimethylsilyl)oxy]-1-(toluene-
4-sulfonyl)-1H-indole-3-carbaldehyde (3.4)
TsCl (18.4 mg, 0.0965 mmol) was added in one portion to
a stirred solution of crude 3.3 (from above experiment and as-
sumed to represent 0.0794 mmol) and Et3N (0.03 mL, 0.2 mmol) in
DMF (2 mL), and stirring was continued for 4 h (Ar atmosphere).
The mixture was extracted with Et2O (10 mL) and the organic ex-
tract was washed with water and saturated aqueous Na2CO3, dried
(Na2SO4) and evaporated. The oily residue was pure enough for the
next step and its use led to the nitrovinyl indole 3.5 in 53% over
three steps from 3.2. Compound 3.4 had: FTIR (film cast) 3391,
2956, 2930, 2886, 2859, 1598, 1554, 1458, 1423, 1373, 1263, 1189,
1174, 1111, 1088, 1013, 960, 840, 670 cmꢂ1
;
1H NMR (400 MHz,
CDCl3) 0.26 (s, 6H), 1.06 (s, 9H), 2.39 (s, 3H), 6.95 (apparent dt,
d
J¼0.4, 8.9 Hz, 1H), 7.29e7.31 (m, 1H), 7.80e7.83 (m, 1H), 7.85 (ap-
4.2. 4-Bromo-5-[(tert-butyldimethylsilyl)oxy]-1H-indole (3.2)
parent dd, J¼0.4, 8.9 Hz, 1H), 8.35 (apparent q, J¼0.4 Hz, 1H); 13C
NMR (100 MHz, CDCl3)
d
ꢂ4.1 (q), 21.7 (s), 25.7 (overlapping signals,
Imidazole (26.2 mg, 0.385 mmol) followed by t-BuMe2SiCl
(47.0 mg, 0.312 mmol) were added in single portions to a stirred
solution of 3.1 in CH2Cl2 (3 mL). Stirring was continued overnight
and the mixture was diluted with CH2Cl2 (10 mL), washed once
with water, dried (MgSO4) and evaporated. The residue was then
filtered through flash chromatography silica gel (1ꢁ1 cm), using
20% EtOAcehexanes (10 mL), to afford 3.2 (76.4 mg, 80%) as an oil:
FTIR (film cast) 3423, 2956, 2930, 2886, 2858, 1567, 1471, 1438, 1411,
1278, 1242, 1185, 1076, 1003, 885, 834, 781, 761, 723 cmꢂ1; 1H NMR
both q), 106.1 (s), 113.3 (d), 118.0 (d), 122.3 (s), 127.3 (d), 128.6 (s),
130.3 (d), 130.6 (s), 132.2 (d), 134.2 (s), 146.2 (s), 150.3 (s), 186.6 (d);
exact mass (electrospray) m/z calcd for C22H2679BrNNaO4SSi
(MþNa) 530.0427, found 530.0427.
4.5. 4-Bromo-5-[(tert-butyldimethylsilyl)oxy]-
3-(2-nitrovinyl)-1-(toluene-4-sulfonyl)-1H-indole (3.5)
NH4OAc (6.2 mg, 0.080 mmol) was added to a stirred solution of
3.4 (from above experiment and assumed to correspond to
0.0794 mmol) in MeNO2 (3.6 mL) and the mixture was heated at
100 ꢀC for 3 h (Ar atmosphere). Evaporation of the solvent and fil-
tration of the residue through a pad of flash chromatography silica
gel (1ꢁ3 cm), using 10% EtOAcehexanes (10 mL), gave 3.5 (23.1 mg,
53% over the three steps from 3.2) as a light orange oil. The material
was unstable and was fully characterized after reduction (see
below).
(400 MHz, CDCl3)
0.8 Hz, 1H), 6.81 (d, J¼8.7 Hz, 1H), 7.19 (dd, J¼8.7, 0.8 Hz, 1H),
7.22e7.23 (m, 1H), 8.14 (br s, 1H); 13C NMR (100 MHz, CDCl3)
d
0.25 (s, 6H), 1.08 (s, 9H), 6.56 (ddd, J¼3.1, 2.3,
d
ꢂ3.9
(q), 18.7 (s), 26.2 (q), 103.5 (d), 105.8 (s), 110.4 (d), 116.4 (d), 125.5
(d), 130.1 (s), 131.3 (s), 146.5 (s); exact mass (EI) m/z calcd for
C14H2079BrNOSi 327.0477, found 327.0476.
4.3. 4-Bromo-5-[(tert-butyldimethylsilyl)oxy]-1H-indole-
3-carbaldehyde (3.3)
In another experiment, using 3.3 (19.8 mg), the nitrovinyl indole
3.5 was obtained in 94% yield over the two steps from 3.3.
POCl3 (10% v/v in DMF solution prepared 5 min before needed,
0.13 mL, 0.139 mmol) was added by syringe to stirred DMF (1.3 mL)
4.6. 4-Bromo-5-[(tert-butyldimethylsilyl)oxy]-
(Ar atmosphere). After 20 min,
a
solution of 3.2 (25.9 mg,
3-(2-nitroethyl)-1-(toluene-4-sulfonyl)-1H-indole (3.6)
0.0794 mmol) in DMF (0.8 mL plus 0.4 mL as a rinse) was added by
syringe, and stirring was continued for 1 h. Saturated aqueous
NaHCO3 was then added dropwise by Pasteur pipette until the
mixture was ca. pH 9 (litmus paper). A slight exotherm was
detected during this quenching. The basic mixture was stirred for
30 min, diluted with water (ca. 10 mL) and extracted with Et2O
(2ꢁ10 mL). The combined organic extracts were washed with water
and brine, dried (Na2SO4) and evaporated. The crude oily product
was used immediately. The material was of satisfactory quality and
the yield of the 3-nitrovinyl indole 3.5 (see later) subsequently
obtained from it was 53% over the three steps from 3.2.
In another experiment, using 3.2 (76.4 mg), flash chromatogra-
phy of the crude product over silica gel (1ꢁ30 cm), using 20%
EtOAcehexanes (50 mL), 40% EtOAc/hexanes (50 mL) and then 50%
EtOAcehexanes (30 mL), gave 3.3 [24.1 mg, 44% after correction for
recovered 3.2 (25.9 mg)] as an analytically pure sample. Aldehyde
LiBH4 (1 M in THF, 0.07 mL, 0.07 mmol) was added by syringe to
a stirred and cooled (0 ꢀC) solution of 3.5 (29.1 mg, 0.0526 mmol) in
THF (2.2 mL) (Ar atmosphere). The mixture was stirred for 30 min
and then quenched by addition of 13 drops (Pasteur pipette) of
aqueous tartaric acid (1 M), followed by water (2 mL). The mixture
was extracted with Et2O (15 mL) and the organic extract was
washed with water and brine, dried (MgSO4) and evaporated. Flash
chromatography of the residue over silica gel (0.5ꢁ10 cm) in
a Pasteur pipette, using 10% EtOAcehexanes, gave 3.6 (28.8 mg,
99%) as an oil: FTIR (film cast) 3108, 2956, 2930, 2896, 2858, 1597,
1533, 1455, 1421, 1375, 1264, 1189, 1174, 1131, 1113, 1088, 1013, 990,
966, 929, 836, 812, 783, 669 cmꢂ1; 1H NMR (400 MHz, CDCl3)
d 0.25
(s, 6H), 1.05 (s, 9H), 2.36 (s, 3H), 3.64 (dt, J¼6.9, 0.7 Hz, 2H), 4.71 (t,
J¼6.9 Hz, 2H), 6.88 (d, J¼8.9 Hz, 1H), 7.24 (apparent dd as part of
a higher order multiplet, J¼8.6, 0.7 Hz, 2H), 7.45 (s, 1H), 7.68

4456
E.J.L. Stoffman, D.L.J. Clive / Tetrahedron 66 (2010) 4452e4461
(apparent d as part of a higher order multiplet, J¼8.4 Hz, 2H), 7.80
4.9. 5-Iodo-6-[(methanesulfonyl)oxy]-2-oxo-2H-chromene-
3-carboxylic acid allyl ester (4.2b)
(d, J¼8.9 Hz, 1H); 13C NMR (100 MHz, CDCl3)
d
ꢂ4.1, 18.3, 21.6, 24.4,
25.7, 76.0, 106.0, 113.3, 117.2, 117.6, 126.8, 127.3, 129.3, 130.0, 131.0,
134.6, 145.3, 149.2; exact mass (electrospray) m/z calcd for
C23H2979BrN2NaO5SSi (MþNa) 575.0642, found 575.0647.
In another experiment, using 3.5 (23.1 mg), a yield of 98% was
obtained.
Allyl alcohol (1.0 mL, 15 mmol) followed by Ti(OPr-i)4 (0.02 mL,
0.07 mmol) were added by syringe to a stirred solution of 4.1
(0.50 g, 1.14 mmol) in PhMe and the mixture was refluxed for 4 h
(N2 atmosphere) and then evaporated. Slow crystallization of the
residue from allyl alcohol (ca. 6 mL), and washing of the crystals
with EtOH (95%) gave 4.2b (0.3391 g, 66%) in one crop. The com-
pound was judged to be pure by 1H NMR, but full characterization
data were not obtained; the structure of 4.2b is established by
conversion into the fully characterized 4.4b (see below): 1H NMR
4.7. N-[2-[4-Bromo-5-[(tert-butyldimethylsilyl)oxy]-
1-(toluene-4-sulfonyl)-1H-indol-3-yl]ethyl]acetamide (3.8)
Zn powder (25 mg, 0.382 mmol) was added to a stirred solution
of 3.6 (9.1 mg, 0.0164 mmol) in 1,4-dioxane (1.3 mL), and AcOH
(0.03 mL, 0.524 mmol) was added by syringe (Ar atmosphere).
Stirring was continued for 4 h and the mixture was filtered through
Celite using MeOH (15 mL). Evaporation of the filtrate gave 3.7 as
a white solid. The 1H NMR spectrum was of poor quality, but it was
not clear if this was due to the presence of impurities, and so the
material was acetylated for characterization.
(300 MHz, CDCl3)
d
3.38 (s, 3H), 4.89 (ddd as an apparent dt, J¼4.7,
1.5, 1.5 Hz, 2H), 5.35 (ddt as an apparent dq, J¼10.6, 1.3 Hz, 1H), 5.50
(ddt as an apparent dq, J¼17.1, 1.5 Hz, 1H), 6.05 (ddt, J¼17.2, 10.6,
5.7 Hz, 1H), 7.40 (dd, J¼9.1, 0.6 Hz, 1H), 7.68 (d, J¼9.1 Hz, 1H), 8.78
(d, J¼0.6 Hz, 1H).
In a larger scale experiment, allyl alcohol (69 mL, 1.0 mol) fol-
lowed by Ti(OPr-i)4 (0.64 mL, 2.2 mmol) were added to a stirred
solution of 4.1 (26.06 g, 59.47 mmol) in PhMe (180 mL). The mix-
ture was stirred and heated at 80 ꢀC for 24 h (Ar atmosphere), at
which time ca. 50% conversion had taken place (1H NMR). A second
portion of Ti(OPr-i)4 (0.66 mL, 2.2 mmol) was added by syringe and
stirring was continued overnight (ca. 12 h) at which time the ob-
served conversion was ca. 80% (1H NMR). Further portions of allyl
alcohol (18 mL, 0.26 mol), followed by Ti(OPr-i)4 (0.60 mL,
2.0 mmol), were added and stirring was continued for ca. 12 h, at
which point the conversion was observed to be >95% (1H NMR).
Volatile material was then removed in vacuo and the crude residue
was used for the conjugate addition to prepare 4.3b (see below),
which was obtained in a yield of 92% over the two steps from 4.1.
Pyridine (10% v/v in Et2O, 0.08 mL, 0.10 mmol) followed by Ac2O
(5% v/v in Et2O, 0.04 mL, 0.02 mmol) were added by syringe to
a stirred and cooled (0 ꢀC) white suspension of the crude amine
(from the above step and assumed to be 0.0164 mmol) in Et2O
(1.3 mL). After 10 min the ice bath was removed and the mixture
was poured into a separatory funnel containing saturated aqueous
NaHCO3 (10 mL). More Et2O (10 mL) was added, the mixture was
shaken for 1 min and the organic phase was washed with water and
brine, dried (Na2SO4) and evaporated. Flash chromatography of the
residue over silica gel (0.5ꢁ10 cm) in a Pasteur pipette, using 80%
EtOAcehexanes, gave 3.8 (8.6 mg, 93% from 3.6) as an oil: FTIR (film
cast) 3291, 3097, 2956, 2930, 2858, 1650, 1554, 1451, 1419, 1372,
1261, 1174, 965, 836, 670 cmꢂ1; 1H NMR (400 MHz, CDCl3)
d 0.24 (s,
4.10. 4-Allyl-5-iodo-6-[(methanesulfonyl)oxy]-
2-oxochroman-3-carboxylic acid 2-(trimethylsilyl)ethyl
ester (4.3a)
6H), 1.05 (s, 9H), 1.97 (s, 3H), 2.36 (s, 3H), 3.15 (apparent dt, J¼0.9,
7.0 Hz, 2H), 3.56e3.61 (m, 2H), 5.50 (br s, 1H), 6.90 (dd, J¼0.4,
8.9 Hz, 1H), 7.23 (apparent dd as part of a higher order multiplet,
J¼0.7, 8.7 Hz, 2H), 7.37 (d, J¼0.4 Hz, 1H), 7.71 (apparent d as part of
a higher order multiplet, J¼8.4 Hz, 2H), 7.83 (d, J¼8.9 Hz, 1H); 13C
LiCl (0.45 g, 11 mmol) was dried in a round-bottomed flask by
heating (heat gun) under oil-pump vacuum for ca. 3 min. CuI
(1.8378 g, 9.650 mmol) was then added and the flask was flushed
with Ar. THF (15 mL) was injected and the mixture was stirred at
room temperature until all solids had dissolved (10 min). The stir-
red solution was cooled to ꢂ78 ꢀC and allylmagnesium bromide
(1 M in Et2O, 9 mL, 9 mmol) was added at a fast dropwise rate by
syringe, followed by Me3SiCl (1.2 mL, 9.4 mmol), which was also
added at a fast dropwise rate. A solution of 4.2a (2.3091 g,
4.524 mmol) in THF (10 mL) was then added at a fast dropwise rate
by syringe. The mixture was stirred for 15 min and then quenched
by addition of saturated aqueous NH4Cl (20 mL). The resulting bi-
phasic mixture was stirred at room temperature open to the at-
mosphere for 1 h and then filtered through Celite to remove
copper-containing solids. The filtrate was diluted with EtOAc,
washed with brine, dried (MgSO4) and evaporated. The residue was
filtered through flash chromatography silica gel (2ꢁ4 cm), using
20% EtOAcehexanes, to give 4.3a (2.4887 g, 99%) as an oil: FTIR
(film cast) 3082, 3032, 2955, 2900, 1787, 1736, 1641, 1596, 1575,
1454, 1414, 1375, 1333, 1251, 1162, 972, 954, 925, 860, 836 cmꢂ1; 1H
NMR (100 MHz, CDCl3)
d
ꢂ3.9, 18.6, 21.7, 23.5, 25.9, 26.6, 40.1, 106.6,
113.1, 117.4, 120.4, 125.9, 126.8, 129.9, 130.0, 131.1, 134.9, 145.1, 149.1,
169.9; exact mass (electrospray) m/z calcd for C25H3379BrN2NaO4SSi
(MþNa) 587.1006, found 587.1002.
4.8. 5-Iodo-6-[(methanesulfonyl)oxy]-2-oxo-2H-chromene-
3-carboxylic acid 2-(trimethylsilyl)ethyl ester (4.2a)
Me3SiCH2CH2OH (6 mL, 41.86 mmol) followed by Ti(OPr-i)4
(0.15 mL, 0.50 mmol) were added by syringe to a stirred solution of
4.1 (2.2631 g, 5.165 mmol) in PhMe (18 mL) and the mixture was
refluxed for 1.5 h (Ar atmosphere). The mixture was evaporated and
the residue was filtered through Florisil (2.5ꢁ3.5 cm), using CH2Cl2.
Evaporation of the solvent gave a residue, which was dissolved in
hot EtOAc (10 mL) and hexanes (30 mL). The mixture was boiled
and more hexanes were added until the boiling mixture remained
cloudy. Enough EtOAc to just clarify the solution (3 mL) was added
and the solution was allowed to cool to room temperature to give
4.2a (2.3091 g, 88%) as white crystals: mp 124e126 ꢀC; FTIR (film
cast) 3084, 3032, 2954, 2898, 1767, 1714, 1612, 1591, 1561, 1458,
1413, 1374, 1359, 1334, 1280, 1248, 1232, 1205, 1174, 986, 967, 934,
NMR (500 MHz, CDCl3)
d
ꢂ0.01 (s, 9H), 0.78e0.87 (m, 2H),
2.06e2.12 (m, 1H), 2.51e2.56 (m, 1H), 3.33 (s, 3H), 3.81 (ddd,
J¼10.9, 4.3, 1.8 Hz, 1H), 3.94 (d, J¼1.8 Hz, 1H), 4.06e4.16 (m, 2H),
5.18 (dd, J¼17.0, 1.0 Hz, 1H), 5.24 (d, J¼10.1 Hz, 1H), 5.81 (dddd,
J¼17.0, 10.3, 8.6, 5.8 Hz, 1H), 7.13 (d, J¼9.0 Hz, 1H), 7.38 (d, J¼8.9 Hz,
841, 798 cmꢂ1 1H NMR (400 MHz, CDCl3)
; d 0.10 (s, 9H), 1.17e1.21
(m, 2H), 3.38 (s, 3H), 4.47e4.51 (m, 2H), 7.39 (dd, J¼9.1, 0.7 Hz, 1H),
7.67 (d, J¼9.2 Hz, 1H), 8.72 (d, J¼0.7 Hz, 1H); 13C NMR (125 MHz,
1H); 13C NMR (100 MHz, CDCl3)
d
ꢂ1.3 (q), 17.6 (t), 36.7 (t), 39.8 (q),
CDCl3)
d
ꢂ1.4 (q), 17.5 (t), 39.7 (q), 65.0 (t), 95.8 (s), 118.3 (d), 120.8
44.7 (d), 50.1 (d), 65.5 (t), 95.8 (s), 118.6 (d), 120.6 (s), 122.7 (d),
130.0 (t),132.6 (d),146.5 (s),148.8 (s),163.0 (s),166.7 (s); exact mass
(electrospray) m/z calcd for C19H25INaO7SSi (MþNa) 575.0027,
found 575.0024.
(s), 122.1 (s), 127.5 (d), 146.7 (s), 151.1 (d), 153.3 (s), 155.5 (s), 162.5
(s); exact mass (electrospray) m/z calcd for C16H19INaO7SSi (MþNa)
532.9558, found 532.9558.

E.J.L. Stoffman, D.L.J. Clive / Tetrahedron 66 (2010) 4452e4461
4457
4.11. 4-Allyl-5-iodo-6-[(methanesulfonyl)oxy]-
2-oxochroman-3-carboxylic acid allyl ester (4.3b)
(s), 166.8 (s); exact mass (EI) m/z calcd for C13H13IO5S 407.9529,
found 407.9525.
LiCl (6.7 g, 158.1 mmol) was dried in a round-bottomed flask by
heating (heat gun) under oil-pump vacuum for ca. 7 min. CuI (25.1 g,
131.8 mmol) was then added and the flask was flushed with Ar. THF
(400 mL) was injected and the mixture was stirred at room tem-
perature until all solids had dissolved (10 min). The solution was
then cooled to ꢂ78 ꢀC and allylmagnesium chloride (2 M in THF,
60 mL, 120 mmol) was added at a fast dropwise rate by syringe,
followed by Me3SiCl (15 mL, 138.1 mmol), which was also added at
a fast dropwise rate. A solution of 4.2b [all the material obtained
from 4.1 (26.06 g), assumed to comprise 59.47 mmol, see above] in
THF (100 mL) was then added at a fast dropwise rate by syringe. The
mixture was stirred for 15 min and then quenched by addition of
saturated aqueous NH4Cl (200 mL). The resulting biphasic mixture
was stirred at room temperature open to the atmosphere for 1 h and
then filtered through Celite to remove copper-containing solids. The
filtrate was diluted with EtOAc, washed with brine, dried (MgSO4)
and evaporated. The residue was filtered through flash chromato-
graphy silica gel (5ꢁ10 cm), using 40% EtOAcehexanes, to give 4.3b
(26.25 g, 92% from 4.1) as an oil. The compound was sufficiently pure
for the next step (1H NMR) but full characterization data were not
obtained; the structure of 4.3b is established by conversion into the
fully characterized 4.4b (see below).
4.14. Methanesulfonic acid 4-allyl-5-iodo-2-oxochroman-6-yl
ester (4.4b)
AcOH (1.8 mL, 31 mmol) was added by syringe to a stirred so-
lution of 4.3b (6.76 g, 14.1 mmol) in CH2Cl2 (54 mL). Ph3P (0.1830 g,
0.6977 mmol) followed by Pd(PPh3)4 (0.5361 g, 0.4640 mmol) were
then added in single portions and the mixture was refluxed for 7 h
(Ar atmosphere), at which point the 1H NMR spectrum of an aliquot
(after evaporation of the solvent) indicated 97% conversion. The
reaction mixture was evaporated and an attempt was made to
isolate 4.4b by flash chromatography using EtOAcehexanes. How-
ever, this was not effective, as the material precipitated on the
column, and use of CH2Cl2 as eluent gave impure product. The
material recovered from the chromatography was then crystallized
by dissolving it in a small volume of boiling EtOAc, adding n-hep-
tane until permanent cloudiness developed, and then adding the
minimum amount of EtOAc to produce a clear solution. Some red
oil separated out on heating and this was removed by decanting the
supernatant solution, which was then allowed to cool slowly to give
4.4b (4.48 g, 78%) as white needles.
In a subsequent larger scale experiment we found that the red
oil contains a significant amount of 4.4b and that the crude material
can be filtered through Florisil and then crystallized from i-Pr2O.
Compound 4.4b also crystallizes from isooctane, but a large volume
is required to dissolve the molten crude material. Alternatively,
pure product can be obtained by flash chromatography over silica
gel using 5e10% EtOAcebenzene.
4.12. 4-Allyl-6-hydroxy-5-iodochroman-2-one (4.4a)
AcOH (0.5 mL, 9 mmol) followed by Bu4NF (1 M in THF, 16 mL,
16 mmol) were added by syringe to a stirred solution of 4.3a
(2.4811 g, 4.491 mmol) in DMF (30 mL) and the mixture was then
heated at 100 ꢀC for 4 h, cooled, diluted with EtOAc (60 mL),
washed four times with water and once with brine. The organic
extract was dried (Na2SO4) and evaporated. Filtration of the residue
through flash chromatography silica gel (2ꢁ4 cm), using CH2Cl2,
gave 4.4a (1.3322 g, 96%). The compound was judged to be pure by
1H NMR but full characterization data were not obtained; the
structure of 4.4a is established by conversion into the fully char-
acterized 4.4b (see below). Compound 4.4a had: 1H NMR
4.15. Methanesulfonic acid 3-[(1-carbamoylmethyl)but-
3-enyl]-4-hydroxy-2-iodophenyl ester (4.5)
A three-necked round-bottomed flask was charged with 4.4b
(0.6112 g, 1.500 mmol) and THF (24 mL). The flask was fitted with
a drying tube containing NaOH pellets, a stopper and an adapter
carrying a Pasteur pipette that extended 1 cm below the surface of
the solution. The pipette was connected by Tygon tubing to another
flask containing liquid NH3 as a source of gaseous NH3, which was
bubbled through the THF solution for 2 h. The reaction flask was
then stoppered and stirring was continued until reaction was
complete (TLC control, ca. 4 h). The mixture was evaporated and the
residue was loaded directly onto a pad of flash chromatography
silica gel (1.5ꢁ4 cm), using CH2Cl2. A faster-eluting impurity was
removed by washing the pad with 50% EtOAc/hexanes (TLC con-
trol), and elution with 8% MeOH in CH2Cl2 then gave 4.5 (0.5802 g,
91%): FTIR (film cast) 3462, 3371, 3198, 3078, 3031, 2935, 1662,
1606, 1581, 1419, 1359, 1280, 1254, 1214, 1172, 1129, 969, 918, 853,
(400 MHz, CDCl3) d 2.09e2.18 (m, 1H), 2.41e2.47 (m, 1H), 2.71 (ddd,
J¼1.0, 6.3, 16.2 Hz, 1H), 2.95 (dd, J¼1.8, 16.2 Hz, 1H), 3.23 (dddd,
J¼1.8, 4.1, 6.1, 6.1 Hz, 1H), 5.14e5.17 (m, 1H), 5.186e5.191 (m, 1H),
5.33 (s, 1H), 5.73e5.83 (m, 1H), 6.93 (d, J¼8.9 Hz, 1H), 7.00 (d,
J¼8.9 Hz, 1H).
4.13. Methanesulfonic acid 4-allyl-5-iodo-2-oxochroman-6-yl
ester (4.4b)
Et3N (0.78 mL, 5.60 mmol) followed by MsCl (0.36 mL,
4.7 mmol) were added dropwise by syringe to a stirred and cooled
(0 ꢀC) solution of 4.4a (1.3322 g, 4.2960 mmol) in CH2Cl2 (18 mL).
After the addition the ice bath was removed and stirring was
continued for 30 min. The mixture was washed with water, dried
(MgSO4) and evaporated. Flash chromatography of the residue over
silica gel (1.5ꢁ35 cm), using 25e30% EtOAcehexanes, gave 4.4b
(1.2672 g, 72%): mp 110 ꢀC; FTIR (film cast) 3080, 3027, 2980, 2917,
2849, 1778, 1640, 1595, 1573, 1453, 1412, 1370, 1239, 1219, 1174,
1150,1054, 953, 919, 829, 799, 728 cmꢂ1; 1H NMR (500 MHz, CDCl3)
823, 792, 698 cmꢂ1 1H NMR (400 MHz, acetone-d6)
; d 2.47e2.54
(m, 1H), 2.73e2.81 (m, 1H), 2.78 (dd overlapping with previous
signal, J¼7.2, 2.5 Hz, 2H), 3.29 (s, 3H), 4.00 (dddd, J¼9.1, 6.9, 6.9,
6.9 Hz, 1H), 4.80 (ddt, J¼10.1, 2.2, 1.0 Hz, 1H), 4.91 (ddt, J¼17.1, 2.3,
1.4 Hz, 1H), 5.72 (dddd, J¼17.3, 10.2, 7.3, 7.3 Hz, 1H), 6.22 (br s, 1H),
6.78 (br s, 1H), 6.85 (d, J¼8.8 Hz, 1H), 7.12 (d, J¼8.9 Hz, 1H), 9.37 (br
s, 1H); 13C NMR (125 MHz, acetone-d6)
d 37.4 (t), 39.0 (d and t co-
incident), 47.6 (d), 101.7 (s), 116.1 (t), 117.5 (d), 121.5 (d), 135.3 (s),
137.9 (d), 143.7 (s), 155.0 (s), 174.6 (s); exact mass (electrospray) m/z
calcd for C13H16INNaO5S (MþNa) 447.9686, found 447.9687.
d
2.10e2.16 (m, 1H), 2.44e2.49 (m, 1H), 2.73 (ddd, J¼16.3, 6.3,
1.1 Hz, 1H), 2.99 (dd, J¼16.2, 1.7 Hz, 1H), 3.34 (s, 3H), 3.36 (dddd
partially overlapping with s at
3.34 ppm, J¼6.0, 4.1, 4.1, 1.7 Hz, 1H),
5.16 (dddd as an apparent dq, J¼16.9, 1.3, 1.3, 1.3 Hz, 1H), 5.18e5.20
(m overlapping with signal at
5.16 ppm, 1H), 5.78 (dddd, J¼16.7,
10.2, 8.2, 6.2 Hz, 1H), 7.11 (d, J¼8.9 Hz, 1H), 7.36 (d, J¼9.0 Hz, 1H);
13C NMR (125 MHz, CDCl3)
32.3 (t), 37.0 (t), 39.5 (q), 40.4 (d), 95. 5
(s), 118.5 (d), 119.8 (t), 122.2 (d), 131.6 (s), 132.9 (d), 146.1 (s), 149.4
d
4.16. Methanesulfonic acid 4-allyloxy-
3-[(1-carbamoylmethyl)but-3-enyl]-2-iodophenyl ester (4.6)
d
K2CO3 (2.26 g, 16.4 mmol) was added to a stirred solution of 4.5
(1.2927 g, 3.040 mmol) in 2-butanone (18 mL). Allyl bromide
(0.47 mL, 5.4 mmol) was injected and the mixture was then heated
d

4458
E.J.L. Stoffman, D.L.J. Clive / Tetrahedron 66 (2010) 4452e4461
at 65 ꢀC for 4.5 h, and a second portion of allyl bromide (0.47 mL,
5.4 mmol) was added (Ar atmosphere). After a further 2 h, a third
portion of allyl bromide (0.18 ml, 2.1 mmol) was added and stirring
at 65 ꢀC was continued for 1 h. The mixture was cooled and parti-
tioned between Et2O (50 mL) and water (50 mL). The aqueous
phase was extracted with Et2O (20 mL) and the combined organic
extracts were washed with saturated aqueous NaHCO3 and dried
(Na2SO4). Evaporation of the solvent gave 4.6 (1.3505 g, 95%) as
a white solid: mp 110e111 ꢀC; FTIR (film cast) 3461, 3377, 3192,
3078, 2978, 2934, 1672, 1611, 1582, 1451, 1424, 1365, 1265, 1216,
818, 805, 734 cmꢂ1 1H NMR (300 MHz, CDCl3)
; d 1.40 (s, 9H),
2.45e2.52 (m, 1H), 2.64e2.70 (m, 1H), 3.56 (apparent br s, 3H),
4.45e4.57 (m overlapping with br s, 3H), 4.89 (apparent d, J¼9.9 Hz,
1H), 4.97 (apparentd, J¼17.0 Hz,1H), 5.28(br s, <1Hdueto exchange),
5.30 (apparent dt, J¼10.5,1.4 Hz,1H), 5.40 (apparent dt, J¼17.3,1.5 Hz,
1H), 5.71 (dddd, J¼17.1, 9.8, 7.1, 7.1 Hz, 1H), 6.05 (ddt, J¼17.2, 10.5,
5.3 Hz, 1H), 6.79 (d, J¼8.9 Hz, 1H), 6.88 (d, J¼8.8 Hz, 1H); 13C NMR
(125 MHz, CDCl3) d 28.5 (q), 35.6 (t), 43.1 (t), 51.8 (d), 69.8 (t), 79.0 (s),
98.5 (s),112.8 (d),113.8 (d),116.1 (t),117.8 (t),132.7 (s),133.1 (d),136.5
(d), 149.1 (s), 150.5 (s), 155.9 (s); exact mass (electrospray) m/z calcd
for C19H26INNaO4 (MþNa) 482.0799, found 482.0797.
1173, 1141, 1016, 997, 969, 918, 846, 827, 794, 733 cmꢂ1
;
1H NMR
(400 MHz, CDCl3) 2.47e2.54 (m, 1H), 2.62e2.67 (m, 1H), 2.69 (dd,
d
J¼14.8, 7.0 Hz, 1H), 2.83 (dd, J¼14.7, 7.8 Hz, 1H), 3.25 (s, 3H),
3.97e4.05 (m, 1H), 4.57e4.60 (m, 2H), 4.88e4.96 (two overlapping
m, 2H), 5.32e5.33 (two overlapping br s, 2H), 5.36 (ddt as an ap-
parent dq, J¼10.6, 1.3, 1.3 Hz, 1H), 5.44 (ddt as an apparent dq,
J¼17.2, 1.6, 1.6, 1.6 Hz, 1H), 5.72 (dddd, J¼17.2, 10.0, 7.2, 7.2 Hz, 1H),
6.08 (ddt, J¼17.3, 10.6, 5.4 Hz, 1H), 6.85 (d, J¼9.0 Hz, 1H), 7.28 (d,
4.19. 4-Allyloxy-3-[(1-aminomethyl)but-3-enyl]-
2-iodophenol hydrochloride (4.9)
A solution of HCl in EtOAc (ca. 2.6 M, 8 mL, 20.8 mmol) was
added to a stirred and cooled (0 ꢀC) solution of 4.8 (0.8351 g,
1.8180 mmol) (N2 atmosphere). After the addition the ice bath was
left in place for 5 min and then removed. Stirring was continued for
2 h and the mixture was evaporated. The residue was triturated
under Et2O to give 4.9 (0.7182 g, 99%) as an extremely hygroscopic
solid: FTIR (microscope) 2978, 1641, 1598, 1573, 1480, 1460, 1423,
1365, 1258, 1231, 1189, 1157, 1089, 1019, 996, 919, 810, 736 cmꢂ1; 1H
J¼9.0 Hz, 1H); 13C NMR (100 MHz, CDCl3)
d 37.4 (t), 39.1 (t), 39.1 (q),
47.2 (d), 69.8 (t), 101.2 (s), 112.8 (d), 116.6 (t), 118.6 (t), 121.2 (d),
132.4 (d),136.0 (s),136.2 (d),143.3 (s),155.1 (s),174.0 (s); exact mass
(electrospray) m/z calcd for C16H20INNaO5S (MþNa) 487.9999,
found 488.0002.
NMR (400 MHz, CD3OD)
d 2.50e2.57 (m, 1H), 2.67e2.74 (m, 1H),
4.17. Methanesulfonic acid 4-allyloxy-3-[[1-[(tert-
butoxycarbonyl)amino]methyl]but-3-enyl]-2-iodophenyl
ester (4.7)
3.29e3.33 (m, 1H), 3.55 (dd, J¼12.8, 8.6 Hz, 1H), 3.90 (dddd, J¼8.5,
8.5, 6.5, 6.5 Hz, 1H), 4.49e4.61 (m, 2H), 4.84 (br s, 1H), 4.91 (ap-
parent dd, J¼10.1, 1.8 Hz, 1H), 4.96e5.02 (m, 1H), 5.28e5.31 (m, 1H),
5.41 (ddt as an apparent dq, J¼17.3, 1.6, 1.6 Hz, 1H), 5.72 (dddd,
J¼17.2, 10.1, 7.1, 7.1 Hz, 1H), 6.11 (ddt, J¼17.2, 10.6, 5.4 Hz, 1H), 6.83
(d, J¼8.9 Hz, 1H), 6.91 (d, J¼8.9 Hz, 1H) (the spectrum showed the
presence of an impurity (signals at ca. 1.2 ppm)); 13C NMR
Pb(OAc)4 (10.27 g, 23.16 mmol) was added in one portion to
a stirred and heated solution of 4.6 (10.67 g, 22.93 mmol) in boiling
t-BuOH (120 mL) (bath temperature 125 ꢀC) and stirring was con-
tinued for 30 min (Ar atmosphere). The mixture was cooled to room
temperature and then Celite (ca.10 g) was added and the solids were
filtered off, using CH2Cl2 as a rinse. The filtrate was evaporated and
the residue was filtered through flash chromatography silica gel
(5ꢁ10 cm), using 20% EtOAcehexanes. The second filtration was
necessary to remove lead residues. Evaporation of the solvent gave
4.7 (11.21 g, 91%) as an oil: FTIR (film cast) 3421, 3078, 2977, 2933,
1706, 1509, 1451, 1366, 1250, 1225, 1174, 996, 969, 915, 845, 826,
(125 MHz, CD3OD)
d 36.8, 43.3, 49.7, 71.0, 96.7, 114.9, 115.1, 117.6,
118.4, 131.5, 134.7, 136.7, 151.4, 152.6; exact mass (electrospray) m/z
calcd for C14H19INO2 (amineþH) 360.0455, found 360.0456.
4.20. 3-Allyl-5-[(tert-butyldimethylsilyl)oxy]-4-iodo-1H-
indole (4.12)
CF3CO2H (0.11 mL, 1.5 mmol) was added to a stirred solution of
PhI(OCOCF3)2 (0.6806 g,1.583 mmol) in 1:1 MeCNeH2O (26 mL) and
the mixture was then cooled in an ice bath. A solution of 4.9
(0.5942 g,1.438 mmol) in 1:1 MeCNeH2O (10 mL) was then added at
a fast dropwise rate (over ca. 5 min) (exposure to the atmosphere).
After the addition stirring was continued for 15 min, NaHCO3 (0.43 g,
5.1 mmol) was added and MeCN was evaporated (water pump, room
temperature), leaving an aqueous mixture. The mixture was
extracted with CH2Cl2 (3ꢁ20 mL) and the combined organic extracts
were dried (Na2SO4). The solution (without evaporation) was di-
rectly filtered through flash chromatography silica gel (1.5ꢁ4 cm),
using 35% EtOAcehexanes (80 mL). 2,6-Di-tert-butyl-4-methyl-
phenol (ca. 3 mg) was added to the filtrate, which was then evapo-
rated to give crude quinone imine 4.10 (0.3929 g). This was covered
in PhH (8 mL) and RheC (5% Rh, 26 mg, 0.013 mmol) was added. The
mixture was refluxed for 2 h with vigorous stirring to avoid uneven
heating (Ar atmosphere). It should be noted that the reaction was
still incomplete (TLC control) at this point, but exceeding this re-
action time results in decomposition and a poorer yield. The isom-
erization proceeds to completion in the following protection step.
The mixture was cooled to room temperature and, without evapo-
ration, filtered through flash chromatography silica gel (1.5ꢁ2 cm),
using 40% EtOAcehexanes (50 mL). The filtrate was evaporated and
the residue was dissolved in CH2Cl2 (8 mL). Imidazole (95 mg,
1.4 mmol) followed by t-BuMe2SiCl (168 mg, 1.12 mmol) were added
in single portions and stirring was continued overnight (stoppered
flask but no protection from air). The mixture was washed once with
water, dried (MgSO4) and evaporated. Flash chromatography of the
790 cmꢂ1; 1H NMR (400 MHz, CDCl3)
d 1.39 (s, 9H), 2.50 (apparent
dtt, J¼13.9, 6.9,1.3 Hz,1H), 2.62e2.70(m,1H), 3.26 (s, 3H), 3.56e3.59
(m, 2H), 3.71e3.78 (m, 1H), 4.46 (br s, 1H), 4.53e4.62 (m, 2H), 4.89
(apparent d, J¼9.8 Hz, 1H), 4.95 (apparent dd, J¼17.0, 7.0 Hz, 1H),
5.34 (ddt as an apparent dq, J¼10.6, 1.4, 1.4 Hz, 1H), 5.43 (ddt as an
apparent dq, J¼17.3, 1.6, 1.6 Hz, 1H), 5.71 (dddd, J¼17.2, 10.1, 7.1,
7.1 Hz, 1H), 6.06 (ddt, J¼17.2, 10.6, 5.4 Hz, 1H), 6.85 (d, J¼9.1 Hz, 1H),
7.30 (d, J¼9.0 Hz, 1H); 13C NMR (100 MHz, CDCl3)
d 28.6 (q), 35.5 (t),
39.1 (q), 42.9 (t), 51.6 (d), 69.8 (t), 79.1 (s),101.9 (s),112.6 (d),116.2 (t),
118.5 (t), 121.2 (d), 132.3 (d), 135.0 (s), 136.2 (d), 143.2 (s), 155.1 (s),
155.7 (s); exact mass (electrospray) m/z calcd for C20H28INNaO6S
(MþNa) 560.0574, found 560.0574.
4.18. [2-(6-Allyloxy-3-hydroxy-2-iodophenyl)pent-4-enyl]
carbamic acid tert-butyl ester (4.8)
Benzyltrimethylammonium hydroxide (Triton B, 40% w/w in
MeOH, 3.3 mL, 7.2 mmol) was added to a stirred solution of 4.7
(1.0924 g, 2.0330 mmol) and stirring was continued overnight open
to the atmosphere. The mixture was warmed to 45 ꢀC and stirring at
this temperature was continued for 3 h. The mixture was cooled and
partitioned between water (15 mL) and Et2O (15 mL). The organic
phase was washed with water and brine, dried (Na2SO4) and evapo-
rated. Flash chromatography of the residue over silica gel
(1.5ꢁ30 cm), using 20% EtOAcehexanes, gave 4.8 (0.8544 g, 92%) as
an oil: FTIR (film cast) 3312, 3078, 2978, 2931, 2868, 1682, 1641, 1572,
1515, 1480, 1454, 1423, 1393, 1367, 1278, 1256, 1169, 1020, 996, 914,

E.J.L. Stoffman, D.L.J. Clive / Tetrahedron 66 (2010) 4452e4461
4459
residue over silica gel (1ꢁ35 cm), using 20% EtOAcehexanes, gave
4.12 (0.3954 g, 67% overall) as an oil: FTIR (film cast) 3421, 3075,
2956, 2929, 2895, 2857,1640,1614,1559,1471,1451,1435,1413,1362,
1323, 1280, 1255, 1210, 985, 899, 835, 780 cmꢂ1; 1H NMR (300 MHz,
1H), 3.56 (dd, J¼11.2, 6.7 Hz, 1H), 3.77 (dd, J¼11.2, 2.7 Hz, 1H),
4.10e4.18 (m, 1H), 6.84 (d, J¼8.9 Hz, 1H), 7.41e7.47 (m, 2H),
7.51e7.57 (m, 2H), 7.82e7.85 (overlapping m and s, 2H), 7.87 (d,
J¼8.9 Hz, 1H); 13C NMR (100 MHz, CDCl3)
d
ꢂ3.6, 18.7, 26.2, 29.9,
CDCl3)
d
0.29 (s, 6H),1.09 (s, 9H), 3.87 (dddd as an apparent dq, J¼6.4,
65.9, 71.8, 81.5, 114.4, 115.7, 120.3, 127.0, 127.6, 129.5, 131.1, 133.0,
134.2, 138.0, 152.1; exact mass (electrospray) m/z calcd for
C23H30INNaO5SSi (MþNa) 610.0551, found 610.0548.
1.4,1.4, 1.4 Hz, 2H), 5.06 (ddt as an apparent dq, J¼6.3,1.8, 1.8 Hz,1H),
5.10 (apparent t, J¼1.5 Hz, 1H), 6.11e6.24 (m, 1H), 6.76 (d, J¼8.6 Hz,
1H), 7.00e7.02 (m, 1H), 7.16 (d, J¼8.6 Hz,1H), 7.86 (br s,1H); 13C NMR
(100 MHz, CDCl3)
d
ꢂ3.7,18.6, 26.2, 30.8, 80.2,111.4,114.1,115.4,116.6,
4.24. [1-Benzenesulfonyl-5-[(tert-butyldimethylsilyl)oxy]-
4-iodo-1H-indol-3-yl]-acetaldehyde (5.3)
124.6, 129.0, 132.3, 138.1, 149.0; exact mass (electrospray) m/z calcd
for C17H25INOSi (MþH) 414.0745, found 414.0746.
In an earlier experiment the indole 4.11 was isolated as an oil
and characterized: FTIR (film cast) 3419, 3074, 3000, 2975, 2917,
2849, 1637, 1616, 1563, 1468, 1437, 1413, 1349, 1187, 946, 911, 873,
NaIO4eSiO2 (18% w/w, 1.35 g, 1.14 mmol) was added in one
portion to a stirred solution of 5.2 (0.1861 g, 0.3168 mmol) in
CH2Cl2 (5 mL). Stirring was continued for 30 min by which time
complete and clean conversion to 5.3 was observed by TLC. The
solid was filtered off to afford the crude oily product, which,
without characterization, was subjected to the reductive amination
described below.
793, 753 cmꢂ1 1H NMR (500 MHz, CDCl3)
; d 3.81 (dddd as an ap-
parent dq, J¼6.2, 1.4, 1.4, 1.4 Hz, 2H), 5.06 (ddt as an apparent dq,
J¼17.0,1.7,1.7 Hz,1H), 5.11 (ddt as an apparent dq, J¼10.1, 1.6, 1.6 Hz,
1H), 5.30 (s,1H), 6.16 (ddt, J¼16.4,10.1, 6.1 Hz,1H), 6.93 (d, J¼8.6 Hz,
1H), 7.03 (t, J¼1.3 Hz, 1H), 7.21 (d, J¼8.6 Hz, 1H), 7.94 (br s, 1H); 13C
NMR (100 MHz, CDCl3)
d
30.1, 75.0, 110.5, 112.5, 115.6 (two over-
4.25. [2-[1-Benzenesulfonyl-5-[(tert-butyldimethylsilyl)oxy]-
4-iodo-1H-indol-3-yl]ethyl]benzylamine (5.4)
lapping signals), 124.7, 127.6, 131.8, 137.8, 148.9; exact mass (elec-
trospray) m/z calcd for C11H11INO (MþH) 299.9880, found
299.9882.
AcOH (0.06 mL, 1 mmol) followed by BnNH2 (0.04 mL, 0.4 mmol)
were added to a stirred solution of 5.3 (assumed to be 0.3168 mmol
from above experiment) in 2:1 MeOHeTHF (7.5 mL). NaBH3CN
(26 mg, 0.41 mmol) was then added and stirring was continued
overnight (Ar atmosphere). The solvent was evaporated and the
residue was dissolved in EtOAc (ca. 10 mL) and washed with satu-
rated aqueous NaHCO3 and brine. The organic extract was dried
(MgSO4) and evaporated. Flash chromatography of the residue over
silica gel (1.5ꢁ25 cm), using 5% MeOHe1% Et3N in CH2Cl2, gave 5.4
(0.1483 g, 72%) as an oil. The oily tertiary amine resulting from re-
ductive amination between 5.3 and the secondary amine 5.4 was
also isolated (58.9 mg). Compound 5.4 had: FTIR (film cast) 3063,
3028, 2957, 2927, 2856, 1584, 1551, 1447, 1441, 1413, 1372, 1259, 1175,
4.21. 3-Allyl-5-[(tert-butyldimethylsilyl)oxy]-4-iodo-1H-
indole (4.12)
Imidazole (84.8 mg, 1.246 mmol) followed by t-BuMe2SiCl
(0.1049 g, 0.696 mmol) were added in single portions to a stirred
solution of freshly prepared 4.11 (0.1891 g, 0.6323 mmol) in CH2Cl2.
The flask was stoppered and stirring was continued for ca. 5 h. The
mixture was washed once with water, dried (MgSO4) and evapo-
rated. Flash chromatography of the residue over silica gel
(1.5ꢁ30 cm), using 18% EtOAcehexanes, gave 4.12 (0.2074 g, 79%)
as an oil. The spectral data for this indole are given above.
1157, 1127, 1109, 1089, 1003, 959, 927, 832, 782, 724, 686 cmꢂ1
;
1H
4.22. 3-Allyl-1-benzenesulfonyl-5-[(tert-butyldimethylsilyl)-
oxy]-4-iodo-1H-indole (5.1)
NMR (300 MHz, C6D6) 0.10 (s, 6H), 0.99 (s, 9H), 2.43e2.48 (m, 2H),
d
2.56e2.61 (m, 2H), 3.48 (s, 2H), 6.59e6.71 (m, 3H), 6.86 (dd, J¼8.9,
2.4 Hz,1H), 7.01 (d, J¼2.4 Hz,1H), 7.09 (apparent tt, J¼5.9,1.7 Hz,1H),
7.17e7.25 (m, 3H), 7.39 (s, 1H), 7.68e7.72 (m, 2H), 8.13 (d, J¼8.9 Hz,
Bu4N$HSO4 (ca. 3 mg), PhSO2Cl (0.08 mL, 0.6 mmol) and aque-
ous KOH (50% w/v, 0.1 mL, 0.9 mmol) were added in that order to
a stirred solution of 4.12 (0.2074 g, 0.5018 mmol) in PhH (5 mL). The
flask was stoppered and the mixture was stirred for 1.5 h under air,
diluted with EtOAc (10 mL) and washed with water and brine. The
organic extract was dried (Na2SO4) and evaporated. Flash chro-
matography of the residue over silica gel (1.5ꢁ25 cm), using 6%
EtOAcehexanes, gave material (5.1) containing a slower-running
impurity. The crude sample weighed 0.2980 g and was processed as
described below. The material had: mp 75e79 ꢀC.
1H); 13C NMR (100 MHz, C6D6)
signal at d 25.8 ppm), 48.4, 53.9, 110.0, 115.1, 118.3, 121.7, 124.8, 126.7,
d
ꢂ4.4, 18.3, 25.8, 25.9 (overlaps with
127.1, 128.3 (overlaps with solvent signal), 128.5, 128.9, 131.2, 132.9,
133.1, 139.0, 141.1, 152.5; exact mass (electrospray) m/z calcd for
C29H36IN2O3SSi (MþH) 647.1255, found 647.1266.
The HCl salt of 5.4, prepared by dissolving a sample in 2 M HCl in
EtOAc and then removing the solvent in vacuo, was obtained as
a glass and had: FTIR (film cast) 3403, 2955, 2931, 2896, 2858, 2763,
2615, 1584, 1551, 1471, 1447, 1414, 1373, 1260, 1176, 1159, 1129, 1113,
1089, 1005, 957, 910, 832, 782, 725, 699, 686 cmꢂ1
;
1H NMR
4.23. 3-[1-Benzenesulfonyl-5-[(tert-butyldimethylsilyl)oxy]-
4-iodo-1H-indol-3-yl]propane-1,2-diol (5.2)
(400 MHz, CDCl3) d 0.23 (s, 6H),1.03 (s, 9H), 3.13 (apparent br s, 2H),
3.50 (t, J¼7.9 Hz, 2H), 4.11 (s, 2H), 6.80 (d, J¼8.9 Hz, 1H), 7.22e7.26
(m, 1H), 7.31e7.35 (m, 2H), 7.39e7.43 (m, 2H), 7.49e7.54 (over-
lapping m and s, 2H), 7.59 (apparent d, J¼7.2 Hz, 2H), 7.81 (d,
J¼8.9 Hz, 1H), 7.83e7.86 (m, 2H), 10.24 (br s, 2H); 13C NMR
N-Methylmorpholine-N-oxide (0.2 g, 2 mmol) followed by OsO4
(tiny crystal, catalytic) were added to a stirred solution of 5.1 (as-
sumed to be 0.5018 mmol from the above experiment) in THF
(3 mL) and water (3 mL). The flask was stoppered and covered with
Al foil and the mixture was stirred for 5 h, diluted with EtOAc (ca.
10 mL) and washed with water and brine. The organic extract was
dried (Na2SO4) and evaporated. Flash chromatography of the resi-
due over silica gel (1.5ꢁ20 cm), using 1:1 EtOAcehexanes, gave 5.2
(0.1861 g, 63% over the two steps from 4.12) as an oil: FTIR (film
cast) 3386, 2955, 2930, 2886, 2858, 1584, 1552, 1472, 1463, 1440,
1413, 1371, 1259, 1223, 1175, 1128, 1089, 1004, 960, 909, 832, 816,
(100 MHz, CDCl3)
d
ꢂ3.9,18.4, 22.8, 25.9, 47.3, 51.0, 81.3,113.9,115.6,
118.7, 126.9, 127.4, 129.1, 129.3, 129.4, 130.0, 130.4, 130.5, 132.2,
134.0, 137.8, 151.7; exact mass (electrospray) m/z calcd for
C29H36IN2O3SSi (MþH) 647.1255, found 647.1252.
4.26. 1-Benzenesulfonyl-5-benzyl-6-[(tert-butyldimethylsilyl)-
oxy]-1,3,4,5-tetrahydropyrrolo[4,3,2-de]quinoline (6.1)
A
microwave flask was charged with Pd(PPh3)4 (8.0 mg,
726, 685 cmꢂ1; 1H NMR (300 MHz, CDCl3)
2.14 (br s, 2H), 2.93 (dd, J¼14.7, 8.0 Hz, 1H), 3.30 (dd, J¼14.7, 4.7 Hz,
d
0.28 (s, 6H),1.06 (s, 9H),
0.0069 mmol) and then t-BuONa (10 mg, 0.10 mmol) was weighed
into it in a dry box. The flask was capped with a septum and

4460
E.J.L. Stoffman, D.L.J. Clive / Tetrahedron 66 (2010) 4452e4461
a solution of 5.4 (56.3 mg, 0.0870 mmol) in PhMe (1.5 mL) and Et3N
(0.5 mL) was injected (Ar atmosphere). The mixture was stirred at
200 ꢀC for 18 h, cooled and filtered through Celite (1ꢁ2 cm), using
CH2Cl2. Evaporation of the solvent and flash chromatography of the
residue over silica gel (1ꢁ30 cm), using 6% EtOAcehexanes, gave 6.1
(19.2 mg, 43%) as an oil: FTIR (film cast) 3063, 3029, 2955, 2930,
2857,1613,1573,1495,1472,1448,1438,1368,1350,1251,1174,1122,
0.06 mL, 0.1 mmol) was added and stirring was continued for 4 h.
The mixture was evaporated and the residue was washed with
hexanes to remove non-polar impurities. Attempted extraction of
the product into CH2Cl2 was not successful as the product was not
soluble enough. The residue was therefore combined with the small
amount of material recovered by evaporating the CH2Cl2 extract,
and dissolved in MeOH (ca. 1 mL) with warming. The flask was
stoppered and placed in the freezer (ca. ꢂ10 ꢀC) overnight, during
which time a small amount of 6.4 (5.7 mg, 28%) crystallized. The
mother liquor was evaporated, and the residue ‘A’ was washed with
hexanes at room temperature, with three portions of boiling
CH2Cl2, and then with one portion of boiling CHCl3. The content of
the CH2Cl2 and CHCl3 extracts was analyzed separately by 1H NMR.
The salt 6.4 was found to be insoluble in CH2Cl2 and CHCl3, and
these fractions contained only a trace amount of impure product.
The CH2Cl2 and CHCl3 extracts were combined and added to residue
‘A’. The original crystals from MeOH (5.7 mg) were also added and
all the material was suspended in 95% EtOH. One drop (Pasteur
pipette) of HI was added to ensure that the iodide was obtained,
and the solvent was removed in vacuo. The residue was then dis-
solved, with warming, in 95% EtOH (ca. 0.5 mL), and the solution
was placed in a freezer (ca. ꢂ10 ꢀC) for 3 days. The salt 6.4 (8.6 mg,
43%) crystallized as a grey solid. The mother liquor was concen-
trated in a Craig tube to ca. 0.3 mL and a further crop of (5.2 mg,
26%) of 6.4 was formed after storage in a freezer, bringing the total
yield to 69%: darkens at 229e233 ꢀC, but does not melt; FTIR (film
cast) 3436, 3214, 2553, 1622, 1478, 1424, 1355, 1256, 1229, 1191,
1090, 1000, 964, 839, 781, 724, 686 cmꢂ1
CDCl3)
;
1H NMR (400 MHz,
d
0.21 (s, 6H), 0.93 (s, 9H), 2.50 (dt, J¼5.7, 1.3 Hz, 2H), 3.18 (t,
J¼5.7 Hz, 2H), 4.63 (s, 2H), 6.80 (d, J¼8.6 Hz, 1H), 6.97 (s, 1H),
7.20e7.24 (m, 5H), 7.28 (d overlapping with solvent signal,
J¼8.6 Hz, 1H), 7.41e7.45 (m, 2H), 7.51e7.55 (m, 1H), 7.84e7.87 (m,
2H); 13C NMR (100 MHz, CDCl3, ꢂ60 ꢀC)
ꢂ4.2 (q), 18.2 (t), 20.0 (s),
d
25.7 (q), 47.3 (t), 55.6 (t),104.4 (d),117.6 (d),118.2 (s),119.0 (d),124.2
(s), 126.6 (d), 126.8 (d), 128.0 (d), 128.2 (d), 128.5 (s), 129.1 (d), 131.4
(s), 133.6 (d), 137.2 (s), 139.3 (s), 139.4 (s); exact mass (electrospray)
m/z calcd for C29H35N2O3SSi (MþH) 519.2132, found 519.2131.
4.27. 5-Benzyl-6-[(tert-butyldimethylsilyl)oxy]-
1,3,4,5-tetrahydropyrrolo[4,3,2-de]-quinoline (6.2)
Na2HPO4 (0.2083 g, 1.467 mmol) followed by Na(Hg) (10% Na,
0.26 g, 1.1 mmol) were added in single portions to a stirred solution
of 6.1 (42.2 mg, 0.0814 mmol) in dry MeOH (3.6 mL) and stirring was
continued for 1.5 h (Ar atmosphere). The mixture was then poured
into water (15 ml) and extracted with Et2O (2ꢁ8 mL). The combined
organic extracts were washed once with brine, dried (MgSO4) and
evaporated to afford pure 6.2 (28.9 mg, 98%) as an oil: FTIR (film cast)
3409, 3062, 3027, 2955, 2929, 2896, 2857, 1604, 1503, 1452, 1443,
1360, 1340, 1274, 1253, 1232, 1166, 1067, 990, 839, 779, 700 cmꢂ1; 1H
1120, 1103, 1086, 986, 824, 801 cmꢂ1 1H NMR (400 MHz, CD3OD)
;
d
3.26 (t, J¼5.7 Hz, 2H), 3.71 (s, 6H), 3.93 (t, J¼5.7 Hz, 2H), 6.70 (d,
J¼8.7 Hz,1H), 6.99 (s,1H), 7.17 (d, J¼8.7 Hz,1H); 13C NMR (100 MHz,
NMR (400 MHz, C6D6)
d
0.28 (s, 6H), 1.05 (s, 9H), 2.50 (dt, J¼5.5,
CD3OD) d 20.0, 54.0, 69.6, 104.6, 115.0, 118.9, 120.6, 121.1, 122.5,
1.1 Hz, 2H), 3.20 (t, J¼5.5 Hz, 2H), 4.77 (s, 2H), 6.18 (apparent d,
J¼1.9 Hz, 1H), 6.45 (br s, 1H), 6.62 (d, J¼8.4 Hz, 1H), 6.99 (d, J¼8.4 Hz,
1H), 7.02e7.10 (m, 3H), 7.31e7.33 (m, 2H); 13C NMR (100 MHz, C6D6)
128.9, 149.0; exact mass (electrospray) m/z calcd for C12H14N2NaO
(MþNa) 225.0998, found 225.1002.
(a high quality 13C NMR spectrum could not be obtained)
d
ꢂ3.9 (q),
Acknowledgements
18.5 (s), 21.1 (t), 26.2 (q), 49.1 (t), 56.8 (t),102.5 (d),111.0 (s),116.4 (d),
118.5 (d), 122.7 (s), 126.8 (d), 128.3 (d), 128.4 (d), 130.9 (s), 131.3 (s),
136.0 (s), 140.8 (s); exact mass (electrospray) m/z calcd for
C23H31N2OSi (MþH) 379.2200, found 379.2201.
We thank the Natural Sciences and Engineering Research
Council of Canada for financial support.
Supplementary data
4.28. 6-[(tert-Butyldimethylsilyl)oxy]-
1,3,4,5-tetrahydropyrrolo[4,3,2-de]quinoline (6.3)
Experimental procedures and copies of NMR spectra. Supple-
mentary data associated with this article can be found, in the online
version, at doi:10.1016/j.tet.2010.04.081.
PdeC (10% Pd, 10 mg, 0.0094 mmol) was added to a stirred so-
lution of 6.2 (28.9 mg, 0.0797 mmol) in EtOH (2.2 mL). The mixture
was degassed by applying house vacuum and filling the flask with
H2. This process was repeated twice more and the flask was fitted
with a balloon filled with H2. The mixture was stirred and warmed
at 45 ꢀC for 15 h and then filtered through Celite (1ꢁ1 cm), using
CH2Cl2. Evaporation of the filtrate gave crude 6.3 (17.7 mg, 77%) as
an oil, which was used directly in the final step. The crude material
References and notes
1. (a) Sanders-Bush, E.; Mayer, S. E. In Goodman and Gilman’s The Pharmacological
Basis of Therapeutics, 10th ed.; Hardman, J. G., Limbird, L. E., Eds.; McGraw-Hill:
New York, NY, 2001; p 269; (b) Glennon, R. A. J. Med. Chem. 1987, 30, 1e12.
2. Hugel, H. M.; Kennaway, D. J. Org. Prep. Proced. Int. 1995, 27, 3e31.
3. cf. (a) Spadoni, G.; Balsamini, C.; Diamantini, G.; Di Giacomo, B.; Tarzia, G.; Mor,
M.; Plazzi, P. V.; Rivara, S.; Lucini, V.; Nonno, R.; Pannacci, M.; Fraschini, F.;
Stankov, B. M. J. Med. Chem. 1997, 40, 1990e2002; (b) Kruse, L. I.; Meyer, M. D. J.
Org. Chem. 1984, 49, 4761e4768; (c) Chang, K.-J.; Kang, B.-N.; Lee, M.-H.; Jeong,
K.-S. J. Am. Chem. Soc. 2005, 127, 12214e12215.
had: 1H NMR (300 MHz, C6D6)
d 0.22 (s, 6H), 1.08 (s, 9H), 2.81 (dt,
J¼5.7, 0.8 Hz, 2H), 3.10 (t, J¼5.7 Hz, 2H), 3.97 (br s, 1H), 6.20 (s, 1H),
6.49 (br s, 1H), 6.49 (d, J¼8.5 Hz, 1H), 6.91 (d, J¼8.5 Hz, 1H).
4. cf. (a) Flaugh, M. E. U.S. Patent 6,022,980, 2000. (b) Kruse, L. I.; Young, R. C.;
Kaumann, A. J. WO 93/00333, 1993. (c) Brown, M. A.; Kerr, M. A. Tetrahedron
Lett. 2001, 42, 983e985; (d) Flaugh, M. E.; Crowell, T. A.; Clemens, J. A.; Sawyer,
B. D. J. Med. Chem. 1979, 22, 63e69; (e) For an example based on a zirconocene-
stabilized benzyne, see: Tidwell, J. H.; Buchwald, S. L. J. Am. Chem. Soc. 1994, 116,
11797e11810.
4.29. 1,3,4,5-Tetrahydro-6-hydroxy-5,5-dimethylpyrrolo-
[4,3,2-de]quinolinium iodide (6.4)
NaHCO3 (12 mg, 0.14 mmol) was added to a stirred solution of
6.3 (17.7 mg, 0.0614 mmol) in dry MeOH (1.8 mL). MeI (10% v/v in
MeOH, 0.13 mL, 0.21 mmol) was then added by syringe and stirring
was continued for 1 h. At this point very little reaction had occurred
(TLC control) and so a second aliquot of MeI (10% v/v in MeOH,
0.10 mL, 0.16 mmol) was added. The flask was then capped with
a glass stopper under Ar and stirring was continued overnight
(13 h). A third aliquot of MeI (10% v/v in MeOH, freshly prepared,
5. Clive, D. L. J.; Stoffman, E. J. L. Org. Biomol. Chem. 2009, 7, 4862e4870.
6. (a) Wieland, H.; Wieland, T. Liebigs Ann. Chem. 1937, 528, 234e246; (b) Marki,
€
F.; Robertson, A. V.; Witkop, B. J. Am. Chem. Soc. 1961, 83, 3341e3342; (c)
Robinson, B.; Smith, G. F.; Jackson, A. H.; Shaw, D.; Frydman, B.; Deulofeu, V.
Proc. Chem. Soc. 1961, 310e311; (d) Ghosal, S.; Dutta, S. K.; Sanyal, A. K.; Bhat-
tacharya, S. K. J. Med. Chem. 1969, 12, 480e483.
7. (a) Use of NaBH4: Simeonov, M. F.; Spassov, S. L.; Bojilova, A.; Ivanov, C.; Ra-
deglia, R. J. Mol. Struct. 1985, 127, 127e134; (b) Use of NaBH4esilicaeCHCl3ei-
PrOH, see: Sinhababu, A. K.; Borchardt, R. T. Tetrahedron Lett. 1983, 24, 227e230.

E.J.L. Stoffman, D.L.J. Clive / Tetrahedron 66 (2010) 4452e4461
4461
8. For one of the few references to the preparation of 4-iodo-5-oxyindoles, see:
Moody, C. J.; Swann, E. J. Chem. Soc., Perkin Trans. 1 1993, 2561e2565.
9. Lipshutz, B. H.; Ellsworth, E. L.; Dimock, S. H.; Smith, R. A. J. J. Am. Chem. Soc.
1990, 112, 4404e4410.
10. cf. (a) Jeffrey, P. D.; McCombie, S. W. J. Org. Chem. 1982, 47, 587e590; (b) Kunz,
H.; Unverzagt, C. Angew. Chem., Int. Ed. Engl. 1984, 23, 436e437.
11. (a) PhI(OCOCF3)2 alone gave the same unsatisfactory result as PhI(OAc)2; we did
not test the later in the present of CF3CO2H. (b) We did not establish if the role
of the acid is simply to protonate the nitrogen. (c) The presence of CF3CO2H can
influence the course of PhI(OAc)2 oxidations: Liang, H.; Ciufolini, M.A. J. Org.
Chem. 2008, 73, 4299e4301.
13. cf. Daumas, M.; Vo-Quang, Y.; Vo-Quang, L.; Le Goffic, F. Synthesis 1989,
64e65.
14. (a) Borch, R. F.; Bernstein, M. D.; Durst, H. D. J. Am. Chem. Soc. 1971, 93,
2897e2904; (b) Cf Abdel-Magid, A. F.; Carson, K. G.; Harris, B. D.; Maryanoff, C.
A.; Shah, R. D. J. Org. Chem. 1996, 61, 3849e3862.
15. (a) Gannon, W. F.; Benigni, J. D.; Suzuki, J.; Daly, J. W. Tetrahedron Lett. 1967, 8,
1531e1533; (b) Peat, A. J.; Buchwald, S. L. J. Am. Chem. Soc. 1996, 118,
1028e1030; (c) We isolated the alkaloid as its iodide; normally it is isolated
from natural sources as the chloride.
16. Guram, A. S.; Rennels, R. A.; Buchwald, S. L. Angew. Chem., Int. Ed. Engl. 1995, 34,
1348e1350.
12. Mann, J.; Barbey, S. Tetrahedron 1995, 51, 12763e12774.
17. Sundberg, R. J.; Cherney, R. J. J. Org. Chem. 1990, 55, 6028e6037.